Tool Condition Monitoring Based on Nonlinear Output Frequency Response Functions and Multivariate Control Chart

Tool condition monitoring(TCM)is a key technology for intelligent manufacturing.The objective is to monitor the tool operation status and detect tool breakage so that the tool can be changed in time to avoid significant damage to workpieces and reduce manufacturing costs.Recently,an innovative TCM approach based on sensor data modelling and model frequency analysis has been proposed.Different from traditional signal feature-based monitoring,the data from sensors are utilized to build a dynamic process model.Then,the nonlinear output frequency response functions,a concept which extends the linear system frequency response function to the nonlinear case,over the frequency range of the tooth passing frequency of the machining process are extracted to reveal tool health conditions.In order to extend the novel sensor data modelling and model frequency analysis to unsupervised condition monitoring of cutting tools,in the present study,a multivariate control chart is proposed for TCM based on the frequency domain properties of machining processes derived from the innovative sensor data modelling and model frequency analysis.The feature dimension is reduced by principal component analysis first.Then the moving average strategy is exploited to generate monitoring variables and overcome the effects of noises.The milling experiments of titanium alloys are conducted to verify the effectiveness of the proposed approach in detecting excessive flank wear of solid carbide end mills.The results demonstrate the advantages of the new approach over conventional TCM techniques and its potential in industrial applications.

AN INTELLIGENT TOOL CONDITION MONITORING SYSTEM USING FUZZY NEURAL NETWORKS

Reliable on line cutting tool conditioning monitoring is an essential feature of automatic machine tool and flexible manufacturing system (FMS) and computer integrated manufacturing system (CIMS). Recently artificial neural networks (ANNs) are used for this purpose in conjunction with suitable sensory systems. The present work in Norwegian University of Science and Technology (NTNU) uses back propagation neural networks (BP) and fuzzy neural networks (FNN) to process the cutting tool state data measured with force and acoustic emission (AE) sensors, and implements a valuable on line tool condition monitoring system using the ANNs. Different ANN structures are designed and investigated to estimate the tool wear state based on the fusion of acoustic emission and force signals. Finally, four case studies are introduced for the sensing and ANN processing of the tool wear states and the failures of the tool with practical experiment examples. The results indicate that a tool wear identification system can be achieved using the sensors integration with ANNs, and that ANNs provide a very effective method of implementing sensor integration for on line monitoring of tool wear states and abnormalities.

Method of Monitoring Wearing and Breakage States of Cutting Tools Based on Mahalanobis Distance Features

The Mahalanobis distance features proposed by P.C.Mahalanobis, an Indian statistician, can be used in an automatic on-line cutting tool condition monitoring process based on digital image processing. In this paper, a new method of obtaining Mahalanobis distance features from a tool image is proposed. The key of calculating Mahalanobis distance is appropriately dividing the object into several component sets. Firstly, a technique is proposed that can automatically divide the component groups for calculating Mahalanobis distance based on the gray level of wearing or breakage regions in a tool image. The wearing region can be divided into high gray level component group and the tool-blade into low one. Then, the relation between Mahalanobis distance features of component groups and tool conditions is investigated. The results indicate that the high brightness region on the flank surface of the turning tool will change with its abrasion change and if the tool is heavily abraded, the area of high brightness will increase apparently. The Mahalanobis distance features of high gray level component group are related with wearing state of tool and low gray level component group correlated with breakage of tool. The experimental results show that the abrasion of the tool’s flank surface affected the Mahalanobis distances of high brightness component of the tool and the pixels of high brightness component set. Compared with the changes of them, we found that the Mahalanobis distance of high brightness component of the tool was more sensitive to the abrasion of cutting tool than the area of high brightness component set of the tool. Here we found that the relative changing rate of the area of high brightness component set was not quite obvious and it was ranging from 2% to 15%, while the relative changing rate of the Mahalanobis distance in table 1 ranges from 13.9% to 47%. It is 3 times higher than the changing rate of the area.

BP-Neural-Network-Based Tool Wear Monitoring by Using Wav elet Decomposition of the Power Spectrum

In a drilling process, the power spectrum of the drilling force is related tothe tool wear and is widely applied in the monitoring of tool wear. But the feature extraction andidentification of the power spectrum have always been an unresolved difficult problem. This papersolves it through decomposition of the power spectrum in multilayers using wavelet transform andextraction of the low frequency decomposition coefficient as the envelope information of the powerspectrum. Intelligent identification of the tool wear status is achieved in the drilling processthrough fusing the wavelet decomposition coefficient of the power spectrum by using a BP (BackPropagation) neural network. The experimental results show that the features of the power spectrumcan be extracted efficiently through this method, and the trained neural networks show highidentification precision and the ability of extension.

On-line Tool Condition Monitoring with Improved Fuzzy Neural Network

This paper presents an investigation of tool condition monitoring based on fuzzy neural network for drilling. In this study, five monitoring feature parameters, which will be used to monitor tool condition, are selected by means of vibration signal spectral analysis. In order to meet the need of the system real time, this paper presents a neural network with fuzzy inference. Fuzzy neural network requires less computation than backpropagation neural network, and can easily describe the relationship between the tool conditions and the monitoring indices. The experimental results indicate that the use of vibration signal for on--line drilling condition monitoring is feasible.

Tool Wear Monitoring in Drilling Using Multiple Feature Fusion of the Cutting Force

This paper presents a tool wear monitoring method in drilling process using cutting force signal. The kurtosis coefficient and the energy of a special frequency band of cutting force signals were taken as the signal features of tool wear as well as the mean value and the standard deviation from the time and frequency domain. The relationships between the signal feature and tool wear were discussed; then the vectors constituted of the signal features were input to the artificial neural network for fusion in order to realize intelligent identification of tool wear. The experimental results show that the artificial neural network can realize fusion of multiple features effectively, but the identification precision and the extending ability are not ideal owing to the relationship between the features and the tool wear being fuzzy and not certain.

A Modification of SOM Network(SLFM) and Its Applications in Tool Wear Monitoring and Quality Control

In this paper, a Supervised Linear Feature Mapping(SLFM) algorithm, as a modification of the Kohonen Self Organizing Mapping (SOM),is proposed. The applications in cutting tool wear estimation and quality control and the comparison with a back propagation (BP) algorithm are discussed. The results show that the SLFM algorithm requires less training time and has higher accuracy compared with the BP algorithm. It might be a great potential approach to integrate multi sensor information in process control.

Precise measurement of geometric and physical quantities in cutting tools inspection and condition monitoring: A review

As one of the most important terminals in machining, cutting tools have been widely used for components manufacturing in aerospace and other industries. The quality of these components and processing efficiency are closely linked to the performance of cutting tools. Therefore, it is essential and critical to inspect the cutting tools and monitor the condition during the stage of manufacturing and machining. This review aims to discuss and summarize the key problems, methods,and techniques from the perspective of the tool geometric and the physical quantities measurement,including machine vision, physical sensors and data processing. It is worth mentioning that we focus on the topic of precision measurement methods and discuss universal solutions by identifying the common characteristics of the measured quantities. Eventually, the challenges and future trends for the development of in-depth research and practical applications are concluded. The research and application of precise measurement techniques for geometric and physical quantities will better promote the development of intelligent manufacturing.

APPLICATION OF ACOUSTIC EMISSION SENSOR IN DETECTING TOOL BREAKAGE IN MICRO-MILLING

Acoustic emission （AE） sensors are used to monitor tool conditions in micro-milling operations. Together with the microphone, the AE sensor can detect the tool breakage more accurately and more effectively by applying the wavelet analysis. The processed tool breakage technique by AE sensor is used to perform the wavelet analysis on the experimental data. Results indicate the feasibility of using the AE signals for monitoring the tool condition in micro-milling.

Single Point Cutting Tool Fault Diagnosis in Turning Operation Using Reduced Error Pruning Tree Classifier

Tool wear is inevitable in daily machining process since metal cutting process involves the chip rubbing the tool surface after it has been cut by the tool edge.Tool wear dominantly influences the deterioration of surface finish,geometric and dimensional tolerances of the workpiece.Moreover,for complete utilization of cutting tools and reduction of machine downtime during the machining process,it becomes necessary to understand the develop-ment of tool wear and predict its status before happening.In this study,tool condition monitoring system was used to monitor the behavior of a single point cutting tool to predict flank wear.A uniaxial accelerometer was attached to a single point cutting tool under study.The accelerometer acquires vibrational signals during turning operation on a lathe machine.The acquired signals were then used to extract statistical features such as standard error,variance,skewness,etc.The substantial features were recognized to reduce the utilization of computing resources.They were used to classify the signals as good and three different measures of flank wear by a decision tree algorithm.Frequency domain features were also extracted and shown to be less effective in classification in comparison to statistical features.REPTree(Reduced Error Pruning Tree)algorithm was used in this study.REPTree decision tree algorithm achieved a maximum classification accuracy of 72.77%for all signals combined.When spindle speed and feed rate are considered as the variables the accuracy is about 86.25%.When spindle speed is the only variable parameter the accuracy is about 82.71%.When depth of cut,feed rate and speed of the spindle are considered as variable parameters,the accuracy of the decision tree is around 93.51%.This study demonstrates the performance of REPTree classifier in tool condition monitoring.It can be utilized for tool wear identification and thus improve surface finish,dimensional accuracy of the work piece and reduce machine down-time.Any additional research on the work may involve analysis of different classifier algorithms which could potentially predict tool wear with greater accuracy.

Real-time tool condition monitoring method based on in situ temperature measurement and artificial neural network in turning

Tool failures in machining processes often cause severe damages of workpieces and lead to large quantities of loss,making tool condition monitoring an important,urgent issue.However,problems such as practicability still remain in actual machining.Here,a real-time tool condition monitoring method integrated in an in situ fiber optic temperature measuring apparatus is proposed.A thermal simulation is conducted to investigate how the fluctuating cutting heats affect the measuring temperatures,and an intermittent cutting experiment is carried out,verifying that the apparatus can capture the rapid but slight temperature undulations.Fourier transform is carried out.The spectrum features are then selected and input into the artificial neural network for classification,and a caution is given if the tool is worn.A learning rate adaption algorithm is introduced,greatly reducing the dependence on initial parameters,making training convenient and flexible.The accuracy stays 90%and higher in variable argument processes.Furthermore,an application program with a graphical user interface is constructed to present real-time results,confirming the practicality.

Vibration signal-based chatter identification for milling of thin-walled structure

In high speed milling aeronautical part,tool condition monitoring(TCM)is very important,because it is prone to get a chatter owing to the low stiffness of thin-walled structures.And the TCM is key technology for automated machining.In this paper,aiming to chatter monitoring in thin-walled structure milling,a variational mode decomposition–energy distribution(VMD-ED)method is proposed to improve the identification accuracy.And a moving average root mean square–mean value(MARMS-MV)identification method and a variational mode decomposition–energy entropy(VMD-EE)identification method are also tested.Identification accuracy and computing time of the three methods are compared.The vibration signals collected from the spindle and worktable are also contrasted.The conducted experimental study shows that,the proposed VMD-ED method offers an identification method for chatter monitoring with greater sensitivity,better stability and less computing time,and mounting the vibration sensor on worktable is better than spindle for a chatter monitoring system.